Research map / Detection and segmentation / CNN features
Fully Rotation-Equivariant Spectral-Spatial Learning for Multispectral Object Detection
Peng Zhang, Tingfa Xu, Shuaihao Han, Jianan Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add rotation-equivariant group convolutions (ablation) | YOLOv8n-style baseline · MODA | mAP: 43.3 -> 47.7 | Improved |
| add SpeIW (ablation) | equivariant baseline · MODA | mAP: 47.7 -> 50.3 | Improved |
| add ReCoW (ablation) | equivariant baseline · MODA | mAP: 47.7 -> 49.6 | Improved |
| add oriented-aware head (ablation) | equivariant baseline · MODA | mAP: 47.7 -> 49.3 | Improved |
| add oriented-aware head and ReCoW to SpeIW (ablation) | equivariant + SpeIW · MODA | mAP: 50.3 -> 54.3 | Improved |
| replace discrete RE-DCN spectral operators with SpeIW (ablation) | FressDet · MODA | mAP: 53.2 -> 54.3 | Improved |
| use SpeIW instead of plain C4-CNN (ablation) | FressDet · MODA | mAP: 52.1 -> 54.3 | Improved |
| use MSI input instead of RGB | FressDet · MODA | mAP: 50.6 -> 54.3 | Improved |
| soft spectral and hard spatial routing instead of soft-soft (ablation) | ReCoW · MODA | mAP: 52.4 -> 54.3 | Improved |
| soft spectral routing instead of hard-hard (ablation) | ReCoW · MODA | mAP: 51.2 -> 54.3 | Improved |
| agreement gate instead of addition fusion (ablation) | ReCoW · MODA | mAP: 52.8 -> 54.3 | Improved |
| use both spectral and spatial branches instead of spectral only (ablation) | ReCoW · MODA | mAP50: 72.3 -> 73.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- WaveMamba — WaveMamba: Wavelet-Driven Mamba Fusion for RGB-Infrared Object Detection (ICCV 2025) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- LESSViT: Robust Hyperspectral Representation Learning under Spectral Configuration Shift (NeurIPS 2026 · similarity 0.86)
- FRED: Towards a Full Rotation-Equivariance in Aerial Image Object Detection (AAAI 2024 · similarity 0.86)
- R2Det: Exploring Relaxed Rotation Equivariance in 2D Object Detection (ICLR 2025 · similarity 0.85)
- ReDiffDet: Rotation-equivariant Diffusion Model for Oriented Object Detection (CVPR 2025 · similarity 0.85)
- S2MAE: A Spatial-Spectral Pretraining Foundation Model for Spectral Remote Sensing Data (CVPR 2024 · similarity 0.84)
- Selective Disk Bispectrum: A Complete and Rotation Invariant Image Descriptor (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
